Startup Financial Model: The 5-Sheet Template Investors Actually Read

Why a startup financial model exists

A financial model is a structured argument for how your inputs turn into outputs. It isn't a prediction of the future — it's a way to put your assumptions out in the open where an investor can change a number and watch what the model does. A good one survives that poking; a fantasy falls apart the moment someone touches a driver.

The 5 sheets

Sheet 1 — Assumptions

This sheet centralizes all numerical drivers, organized into categories:

Growth: New customers per month, conversion rate, churn
Pricing: ARPU or contract value, discount mix, payment terms
Unit economics: CAC, gross margin per unit, payback months
Team: Hires by role and month, fully-loaded salary
Other costs: Tools, infra, marketing, legal, office

Color your assumption cells light blue (formula cells stay black). It's the universal convention, and it lets an investor see at a glance which numbers are inputs they can challenge versus outputs the model computes for them.

A worked unit-economics check

Before you trust a single revenue number, sanity-check the engine underneath it. Investors do this in their head in about thirty seconds, so do it first. Say your Assumptions sheet has:

CAC (fully-loaded sales + marketing to land one customer): $600
ARPU: $100/month, i.e. $1,200/year
Gross margin: 75%, so each customer throws off $75/month in gross profit
Monthly churn: 3%, so an average customer stays ~33 months (1 ÷ 0.03)

Now compute the three numbers that tell you whether the business is real:

CAC payback = CAC ÷ monthly gross profit = $600 ÷ $75 = 8 months. Good — a common bar is under ~12 months for SMB/self-serve, under ~18 for enterprise.
LTV = monthly gross profit × lifetime = $75 × 33 ≈ $2,475.
LTV:CAC = $2,475 ÷ $600 ≈ 4:1. Healthy — the rough rule of thumb is ~3:1; below ~1:1 you lose money on every customer, and far above ~5:1 usually means you're under-investing in growth.

If your model produces an LTV:CAC of 12:1 and a 2-month payback, that's not a great business — it's almost always a great big assumption error (CAC too low, churn too optimistic). The other guardrail worth stating: SaaS gross margin typically lands around 70–80%+. If your model needs 95% margins or sub-1% churn to work, that's the tell an investor will catch.

Sheet 2 — Revenue

Bottom-up construction follows this structure:

New customers (leads × conversion)
Churned customers (existing × monthly churn)
Active customers = previous + new − churned
MRR / monthly revenue = active × ARPU
Expansion (upsells, seat growth if applicable)

Models should extend 24–36 months; longer projections lack credibility.

Model growth bottom-up, not as a magic percentage

The number-one tell of a fake model is constant month-over-month growth compounding for 36 months — "we grow 15% MoM" pasted down a row. Nothing grows at a fixed rate for three years; that curve quietly turns a small startup into a company larger than its entire market by month 30, and every investor has seen the trick. Model growth as a mechanism instead, in one of two ways:

As a channel. New customers come from spend ÷ CAC. You put $10,000 into a channel at a $600 CAC, you get ~16 customers — and as you scale that spend, CAC usually rises (the cheap audience saturates). This forces you to fund growth with real dollars, which is the honest version.
As a decaying growth rate. If you do use a percentage, decay it: 20% MoM early, stepping down toward single digits as the base gets big. Growth as a share of a larger and larger number is the S-curve every real company actually rides.

Either approach beats a flat percentage because it bakes in the truth that growth gets harder, not easier, as you scale.

Churn: one number quietly overstates your revenue

"3% churn" hides two different things, and the difference compounds across 36 months:

Logo churn (customers lost) vs revenue churn (dollars lost). With expansion from your remaining accounts, revenue churn can be far lower than logo churn — and at the best companies it goes negative (net revenue retention above 100%). Model the one that's load-bearing for your story, and don't silently swap between them.
Retention curves flatten; they don't decay in a straight line. Real cohorts lose their worst-fit customers early, then the survivors stick around for years — the curve has a long flat tail. A single constant monthly-churn number applies that early loss rate forever, which understates the lifetime of your best customers and quietly mis-prices LTV. If you have any cohort data, use the actual retention curve rather than one blended percentage.

Sheet 3 — Costs

Two distinct blocks organize expenses:

Headcount: One row per hire showing start month and fully-loaded salary (including ~25% for benefits/taxes)
Non-headcount: Tools, hosting, marketing spend, contractors, legal expenses

Marketing expenditure should correlate with CAC assumptions—if CAC is $200 and you project 50 monthly customers, marketing should approximate $10,000.

Sheet 4 — Cash

This sheet determines operational viability:

Opening cash = previous month's closing cash
+ Cash in = revenue collected (accounting for payment terms)
− Cash out = costs paid (considering vendor terms)
Closing cash = opening + in − out
Runway = closing cash ÷ average monthly burn (last 3 months)

"The single most important number on the entire model is the month your closing cash hits zero."

Sheet 5 — Summary

A single-page overview containing:

24-month MRR and cash charts
Headline metrics: ending ARR, ending burn, runway, hires
Three assumptions driving 80% of outcomes (typically growth rate, churn, CAC)
Sensitivity analysis showing runway impact under reduced growth scenarios

Common mistakes

Compounding a flat MoM growth rate for years — the single biggest tell of a model nobody believes. Model growth as a channel or a decaying rate instead.
Hardcoding numbers inside the revenue/cost sheets instead of pointing them at the Assumptions sheet.
Projecting past ~36 months — past that you're just guessing, and everyone knows it.
Pulling top-down market sizing ("1% of a $50B market") into the revenue formulas instead of building bottom-up.
Skipping payment-terms logic and treating bookings as cash in the bank.
Burying burn and the cash-zero month so they don't show up on the summary.

How investors actually read it

Review Summary sheet for runway and ending ARR
Examine Cash sheet to identify negative cash month
Access Assumptions sheet to challenge three disputed numbers, observing model reactivity
If model withstands scrutiny, open Revenue and Costs sheets for structural verification

A note on tools

"Google Sheets is fine. Excel is fine. The tool doesn't matter — the structure does." Sophisticated planning software proves useful with established data but creates friction at pre-revenue stages.